Generative Artificial Intelligence Academic Student Use Policy

A. Why Foundational Skills Cannot Be Outsourced

Generative AI is already embedded in legal practice and will only become more so. Mitchell Hamline is committed to preparing students to use these tools competently, ethically, and effectively. But competent AI use in law requires something AI itself cannot provide: a lawyer who knows enough to evaluate what the AI produces.

An AI tool can generate a brief, draft a contract, or summarize a case. What it cannot do is tell you whether the analysis is sound, whether the cited cases are good law, whether the strategy is wise, or whether the advice serves the client's actual interests. Those judgments require a trained legal mind. A student who uses AI to do legal thinking before learning to do it themselves is not building that mind — they are borrowing one and will not have it when it matters.

This is the foundational insight behind this policy: you must learn to think like a lawyer before you can responsibly use tools that simulate legal thinking. The restrictions on AI use in certain courses and assignments are not obstacles to your education — they enable the education.

B. The Risk of Shortcuts That Stunt Growth and Learning

AI tools can produce answers that look right. They can draft convincing legal prose, generate plausible issue lists, and summarize case holdings with apparent authority. This very fluency is a danger. A student who consistently offloads legal analysis to AI during law school may never develop the ability to recognize when the AI is wrong — and AI is often wrong, in ways that are not obvious.

Research on learning consistently shows that effortful struggle with difficult material builds durable expertise. The discomfort of working through a hard contracts problem, of drafting a memo derived from your own legal analysis, of finding the flaw in an argument — these are not inefficiencies to be eliminated. They are how legal reasoning develops. AI shortcuts that bypass this struggle do not save students time; they deprive students of the experience that makes them lawyers.

This policy therefore distinguishes between uses of AI that support learning and uses that replace it. Using AI to clarify a concept you are wrestling with, to generate practice problems, or to check your grammar after you have done your own writing — these uses can deepen engagement. Using AI to generate the analysis, write the argument, or answer the question that the assignment asks you to answer — these uses hollow out the learning the assignment was designed to produce.

C. Ethics and Professional Responsibility

Lawyers must navigate rules of professional conduct — cited below by their ABA Model Rule numbers — that directly govern how AI can and cannot be used in practice. (Not all jurisdictions have adopted the Model Rules, but the underlying duties are common to the profession).

• The duty of competence under Rule 1.1 requires lawyers to understand the benefits and risks of relevant technology, which courts and bar authorities have interpreted to include AI tools used to generate legal work product — meaning a lawyer who submits AI-generated analysis without understanding or verifying it may fall short of the standard of care.
• The duty of supervision under Rules 5.1 and 5.3 requires that lawyers exercise appropriate oversight over the work of subordinates and non-lawyer assistants, a principle that extends to AI systems: delegating research, drafting, or analysis to an AI tool does not relieve the supervising lawyer of responsibility for the accuracy and quality of the output.
• The duty of confidentiality under Rule 1.6 prohibits lawyers from disclosing client information without consent, and many publicly available AI platforms collect, retain, or use the data entered into them, creating serious risk when lawyers input client facts, case details, or privileged communications into an AI tool without understanding the platform’s data practices.
• The duty of candor to the tribunal under Rule 3.3 requires that lawyers not make false statements of law or fact to a court, and AI tools are well documented to “hallucinate” — generating plausible but fabricated case citations, statutes, and holdings — making verification of all AI-generated legal authority an ethical obligation, not merely a good practice.
• Finally, the duty to provide competent representation intersects with AI’s known limitations in jurisdiction-specific law, minority legal rules, and rapidly evolving areas of doctrine, where AI output may appear authoritative while being substantially inaccurate. Taken together, these rules establish that AI is a tool a lawyer may use, but never a substitute for the professional judgment, verification, and accountability the rules require the lawyer to provide personally.

Graduating from law school without obtaining the necessary skills and knowledge required to be an effective lawyer because of an over-dependence on AI tools in law school puts you at serious risk of violating these ethical rules.

D. Why This Policy Exists

The legal profession is rapidly integrating generative AI into practice. This policy does not restrict learning; it supports it. Mitchell Hamline wants students to graduate as competent, critical, and ethical users of AI, ready for a profession in which AI tools are already standard. This policy creates a common framework that gives students clear, consistent guidance while preserving full flexibility for individual faculty.

Disclosure under this policy is an educational exercise, not a punitive one. The aim is to help students develop the habits of transparency, verification, and critical judgment that the profession requires.

E. What is Generative and Agentic AI?

Generative AI tools are applications that produce new text, analysis, summaries, or other content in response to a prompt. This policy applies whenever a student uses such a tool to generate, draft, analyze, or summarize academic work.

Agentic AI refers to AI systems that can pursue goals with a degree of autonomy — planning multi-step actions, making decisions, using tools, and adapting to feedback, rather than just responding to a single prompt with a single output.

F. What This Policy Covers

This policy governs the use of generative and agentic artificial intelligence (GenAI) tools — including but not limited to ChatGPT, Claude, Gemini, Copilot, Westlaw AI, Lexis+ AI, Bloomberg Law AI, Grammarly AI tools, Harvey, and similar large language model-based applications — by students in:

• All Mitchell Hamline courses (JD, LLM, experiential courses, certificate programs, and other degree programs);
• Law school-administered co-curricular programs.

G. What This Policy Does Not Cover

This policy is solely intended to govern faculty rules regarding students’ use of AI and is not meant to supersede or replace any other policy on AI use at Mitchell Hamline.

This policy does not govern:

• Faculty use of AI in course preparation, grading, or feedback;
• Administrative or institutional use of AI by Mitchell Hamline;
• AI use by employers or supervising attorneys in externships or field placements;
• Student organizations, law review, journals, or moot court boards — these organizations are encouraged to develop their own AI policies.
• Artificial intelligence that is not generative or agentic. Examples of non-generative and non-agentic AI include spellcheck, Google search engines, and GPS systems.

H. Relationship to Other Rules

This policy supplements, and does not replace, the Mitchell Hamline Student Code of Conduct. Violations of this policy may constitute academic misconduct under the Student Code of Conduct. Professional responsibility rules, including the Minnesota Rules of Professional Conduct, apply independently in clinic and externship settings.

Faculty have full authority, consistent with Mitchell Hamline policy, to set the AI policy for their course regarding student use of AI and may vary that policy by assignment. Faculty may prohibit all AI use, permit specific uses, or encourage AI use as part of the learning objectives. The faculty design the policy that fits their course. Faculty are encouraged to communicate their AI policy clearly in the syllabus and in assignment instructions.

The three-level framework below offers a simple way to signal expectations, although faculty may use another method to explain their AI student use policy.

Faculty policy is the default. These rules apply only to fill gaps, including when a faculty member has not addressed a specific situation in the syllabus or assignment instructions. When in doubt, ask your professor before submitting, not after.

A. AI Is Prohibited by Default for All Assessments Completed for Credit

Unless your professor expressly permits AI use, AI tools may not be used for any examination or assignments completed for credit, even if not graded.

B. AI Is Permitted for These Uses, Except Where Restricted by Section III.A

Unless your professor prohibits it, AI tools may be used for:

• Background research and understanding legal concepts;
• Organizing your thoughts before you write;
• Checking grammar and spelling.

The analysis, reasoning, and writing you submit must be your own.

C. Collaborative and Group Work

This policy applies equally to all members of a collaborative assignment. Specifically:

• If AI use is permitted on a group assignment, the disclosure obligation applies to the group as a whole. The group must collectively disclose any AI use in the submission.
• If group members disagree about whether to use AI on a restricted assignment, they must raise the issue with the professor before submitting the work —not after.
• No individual group member may use AI in violation of the applicable rule.

D. Assistive Technology and Disability Accommodations

This policy does not restrict the use of assistive technology tools that are part of a student's documented disability accommodation. Faculty and students with concerns about an approved use of assistive technology should consult with the Office of Disability Services

Disclosure is an educational practice, not a punishment. Law practice increasingly requires attorneys to disclose AI use to courts and clients. Developing this habit in law school prepares you for practice. Students should follow their professor’s disclosure rules, but the following default rules apply when there are no disclosure instructions in a professor’s syllabus or assignment.

A. Grading Protection

When AI use is permitted on an assignment, a student who discloses that use in accordance with this policy will not receive a lower grade or otherwise be penalized for having used AI. Faculty may not penalize a student for using AI in a manner the faculty has permitted.

B. What to Disclose

When the professor requires disclosure, include the following:

• The name of the AI tool(s) used (e.g., ChatGPT-4o, Claude 3.5 Sonnet, Copilot, Westlaw CoCounsel);
• How you used the tool (e.g., generated an outline, summarized a case, drafted a section);
• How you verified the AI output — what steps you took to check accuracy, identify errors, or confirm citations;
• Whether you agree with the AI output and why, and how the AI output was changed or incorporated in your final work.

Faculty may prescribe a different or more detailed disclosure format. Faculty may require process documentation — such as annotated research files, draft submissions, or a brief process memo — for any assignment. Failure to produce credible process documentation when requested may be considered as relevant evidence in a review under the Student Code of Conduct.

Follow your professor's instructions when provided.

C. Sample Disclosure Statement

Unauthorized AI use — AI use that violates this policy, your professor's instructions, or exam rules — may constitute a violation of the Mitchell Hamline Student Code of Conduct and may be treated as academic misconduct.

Examples of prohibited conduct include:

• Submitting AI-generated text as your own work without disclosure or attribution;
• Using AI during an examination when it is not permitted;
• Using AI to generate the legal analysis or argument on a writing assignment restricted from AI use;
• Falsely representing in a disclosure that you did not use AI.

Suspected violations will be addressed under the Mitchell Hamline Student Code of Conduct. Mitchell Hamline does not rely solely on automated AI-detection software as sufficient evidence of a policy violation. AI-detection tools can produce significant rates of false positives and have documented disparate impacts on non-native English speakers, and do not establish whether a student acted in bad faith. A flag from a detection tool may prompt a faculty-student dialogue (see below) but is not, by itself, grounds for a formal academic misconduct referral.

Relevant evidence in a Student Code of Conduct proceeding may include the following.

• Mitigating factors: presence of an AI Disclosure Statement; a student's ability to explain and defend their submitted work.
• Exacerbating factors: absence of an AI Disclosure Statement; a student's inability to explain and defend their submitted work; inconsistencies between submitted work and the student's authenticated prior writing; absence of a credible process trail when one was required or reasonably expected; use of cases or lines of reasoning that differ from those used in class or the textbook; and the presence of fabricated citations or other hallmarks of AI-generated content.

Where possible, the preferred initial approach is a faculty-student dialogue to understand what occurred before initiating any formal process under the Student Code of Conduct. Faculty may, at their discretion, require any student to meet to discuss, explain, or expand upon any submitted assignment. This applies to any assignment, whether or not AI use is suspected, and is a normal part of legal education. A student's inability to explain or defend their submitted work in a faculty conversation may be considered as relevant evidence in a review under the Student Code of Conduct. Students should expect faculty to question them in individual meetings to ensure they have learned the required material.

Even when AI is permitted, keep these professional principles in mind:

Verify Everything. AI tools frequently produce incorrect, outdated, or fabricated legal citations and analysis. You are responsible for verifying every legal authority you cite. Submitting an AI-hallucinated case or statute is not an excuse — it reflects on your professional competence.

Your Judgment Is Not Optional. A lawyer cannot delegate professional judgment to an AI. AI tools should assist with your work; they should not produce the work product. Critical analysis, ethical reasoning, and advocacy are skills you must develop personally — they cannot be outsourced.

Be Familiar with the Tools. Employers increasingly expect graduates to have broad familiarity with AI tools used in legal practice, including Westlaw AI, Lexis+ AI, Microsoft Copilot, ChatGPT, and others. You need not be an expert in every tool, but you should be aware of what these tools do and how they are used in the profession.

Confidentiality and Privacy. Before using any AI tool, understand its data practices. Information entered into a public AI tool may be stored, used for training, or exposed. Apply the same care you would with any client or sensitive information.

Bias and Limitations. AI tools reflect the biases in their training data. They may perpetuate stereotypes, misstate minority legal rules, or miss jurisdiction-specific nuances. Always apply critical analysis to AI output.

Competence Requires Practice. The skills tested on the bar exam — issue spotting, legal analysis, research, writing — must be developed through practice. Over-reliance on AI in law school may undermine your ability to demonstrate these skills when it matters most.

This policy will be reviewed annually by the Curriculum Committee. The review process will:

• Solicit feedback from students and faculty through the Curriculum Committee Chair;
• Consider developments in AI technology, legal practice, and bar examination requirements;
• Result in a revised policy that is presented to the full faculty for a vote before the start of each academic year.

Students or faculty who wish to submit feedback for the annual review should contact the Curriculum Committee Chair. This policy will be posted on the Mitchell Hamline intranet and updated promptly following any faculty vote.

If you are unsure whether AI use is permitted for a specific assignment:

• Read your syllabus and assignment instructions carefully;
• Ask your professor before submitting — not after;
• Contact the Office of Academic Excellence for guidance on study strategies involving AI;
• For disability accommodation questions related to AI tools, contact the Office of Disability Services.

To submit feedback for the annual policy review, please contact the Curriculum Committee Chair (currently, Prof. Niedwiecki).

Adopted by the Faculty on September 16, 2026, Effective Academic Year 2026-27. (Reviewed Annually by the Curriculum Committee. Faculty vote required before each academic year.)

Disclaimer

Mitchell Hamline policies are for informational purposes and to set expectations and do not constitute a contract, either express or implied, with any employee, student, or third party. Subject to applicable law, Mitchell Hamline reserves the right to amend, modify, or terminate any of the policies or procedures described herein, in whole or in part, at any time, with or without prior notice, at its sole discretion. Mitchell Hamline also may choose not to apply a policy in certain circumstances if it determines a different course of action is more appropriate.